|Publication number||US4069675 A|
|Application number||US 05/667,347|
|Publication date||Jan 24, 1978|
|Filing date||Mar 16, 1976|
|Priority date||Mar 16, 1976|
|Also published as||CA1170750A, CA1170750A1, DE2710353A1|
|Publication number||05667347, 667347, US 4069675 A, US 4069675A, US-A-4069675, US4069675 A, US4069675A|
|Inventors||Louis S. Adler, S. Keith Swanson|
|Original Assignee||Measurex Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (2), Referenced by (40), Classifications (18)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The present invention is directed to a method of optimizing the performance of a multi-unit power plant and more specifically for the integrated control of the power boiler and steam distribution (including turbine generator) system in the pulp and paper industry.
In the pulp and paper industry and for that matter in any petro-chemical industry in-house multi-unit power plants are used both to generate electrical energy and steam which is used as part of the processing operation; for example, in a paper mill in the paper industry. The steam energy is a significant fraction of the total energy produced by the power plant. Moreover, this steam energy is used for purposes independent of generation of electrical power.
This is in contrast to a typical power industry system where all steam from the turbine generator is returned via condensate to the power boiler for reheating along with make-up water. In the paper industry, for example, a significant portion of steam is extracted from the turbine generator at pressures of, for example, 600, 160 and 60 pounds per square inch and used to provide steam energy to many different types of production units, for example, paper machines, dryers, evaporators, in the pulp and paper manufacturing process.
Thus, it is apparent that the control criteria for energy optimization in the typical power industry systems are inadequate in the paper industry. Yet another difference especially in the paper industry is the use of wood chips as one of the fuels for the power boiler in addition to both coal and oil. Wood chips because of the change in their moisture and type of wood will vary in their effective heating value. In addition, this heating value may be lessened because of the tendency of the wood chips to form soot in the power boiler to again lower efficiency. Moreover, the perturbations in fuel characteristics and their resulting effect on efficiency cannot be directly and accurately measured. This is especially true in present day electrical power industry control systems.
There are several specific disadvantages in the case of pulp mills with present energy management systems. For example, in power boilers which are fired with two fuels, for example, coal or wood chips as a base load fuel and oil as the swing fuel, the base load fuel is not adjusted for changes in operating conditions. The loading of individual power boilers is based solely on steam production requirements not on operating efficiency. The efficiencies of the various steam turbines are not taken into account when loadings are distributed to the steam engine. Finally, presently available systems are not able to compensate quickly for changes in electrical and steam demands. The result is higher rate charges from the local electrical utility due to exceeding the peak demand level.
It is, therefore, a general object of the invention to provide a method of optimizing the performance of a multi-unit power plant and more specifically a power plant which has as significant independent outputs low pressures and medium pressure steam for use in the chemical processing portion of the plant.
In accordance with the above object there is provided a method of optimizing the performance of a multi-unit power plant which produces steam energy from a plurality of different fuels. The steam energy used for different purposes is a significant fraction of the total energy produced by the power plant. Input perturbations to the system affecting its performance cannot be directly and accurately measured. The real time incremental efficiency of the units of the power plant are determined by bump tests. Change in the energy output of the units in response to change in energy input is sensed. The index of performance of the units is determined by utilizing the incremental efficiency determination. The energy outputs of the units are reallocated in accordance with the index of performance.
FIGS. 1 and 2 are functional diagrams illustrating power boiler and steam distribution system optimization;
FIG. 3 is a block diagram of a multi-unit power plant utilizing the method of the present invention;
FIG. 4 is a generalized control schematic of a typical turbine generator;
FIGS. 5 and 6 are flow charts useful in understanding one embodiment of the invention;
FIG. 7 is a flow chart useful in understanding another embodiment of the invention; and
FIGS. 8 and 9 are steam flow diagrams useful in understanding the flow charts of FIGS. 5 and 7, respectively.
FIG. 1 is a functional diagram which illustrates power boiler system optimization. As indicated the inputs to the system are the steam demand which would include low and medium pressure steam for the processing portion of the pulp mill, for example, and also the steam demand of the turbine generator (see FIG. 2) for producing the necessary electrical energy. As discussed above, the prime purpose of the electrical energy is to meet peak demands and thus minimize costs of utility power. Other inputs to the block 10 labeled redistribution algorithm are, fuel efficiency or η, the cost of fuel which as will be discussed below provides an index of performance, and constraints on the system would include antipollution controls such as the time and duration of the burning of wood chips. Finally, a feed-forward from the steam distribution system of FIG. 2 provides for more level operation. Once the redistribution algorithm is computed and the allocations are actually determined as indicated in box 11 by receiving the actual steam flows and other operational information such as fuel input flows, both steam targets and fuel targets are set for the power boilers of the system designated PB1, PB2, PB3 . . .
The steam distribution system of FIG.2 is similar in that the redistribution algorithm block 10' has as inputs steam demands, electrical demands, utility input and constraints. One of the constraints here is that peak electrical power requirements of the system are met to minimize utility power costs. Allocation to various turbine generators designated TG1, 2, 3 . . . determines both the electrical output target and the extraction targets which are in effect steam at the various pressures. FIG. 3 illustrates a typical multi-unit power plant which would be used in a paper mill and includes various power boilers 12 designated PB1 through 7 which feed various steam headers 13a through c. These are relatively high pressure, for example, 600 psi. and 400 psi. and primarily provide steam energy to the first turbines indicated 14a through 14f. In addition, there is a high pressure turbine 14g which receives steam from a 1500 psi. header. Low and medium pressure steam is extracted from the various turbine generators to the 150 psi. headers and 60 psi. headers. A typical use of the 150 psi. headers is coupling them directly to a paper machine as indicated on line 16.
A more detailed control schemmatic of a single turbine generator 14' is shown in FIG. 4 where the extraction flows of 600, 160, 60 and essentially zero psi. are indicated. Moreover, a computer performs a typical feedback control. However, in accordance with the invention, a turbine generator load allocator 11' (see FIG. 2) is illustrated which is controlled by the system optimization method of the present invention.
FIG. 5 illustrates an incremental efficiency bump test for a single boiler. The boiler is indicated as having two fuel inputs; for example, oil and coal or wood chips. An input flow valve sets the fuel flow f1 in pounds per hour as measured by the flow meter F and an output flow meter F which measures the steam production in pounds per hour, S1.
As illustrated in FIG. 5, the steps of the test are as follows. To initiate the bump test fuel number one is perturbed a Δf1 amount. Specifically, the amount of oil that enters the boiler may be changed by, for example 5%. The resultant change in steam production S1 is measured and the incremental efficiency calculated by taking the ratio of the two amounts. This determines the incremental efficiency for the fuel number one and fuel one is returned to its original set point. Then fuel number two such as coal or wood chips is perturbed. Measurement of change in input energy with the change in energy output is again made for the incremental efficiency of the second fuel. In the case of "method #1" this bump test is ideally made as illustrated in FIG. 8 after each change in steam flow. This may occur on an hour to hour basis.
After computing or determining the incremental efficiencies these are utilized when there is a change in steam demand, to compute the index of performance. In simplified form if steam production is reduced then this is done with the least efficient boiler and an increase in production is with the most efficient boiler. FIG. 6 illustrates the optimization process after fuel efficiencies are computed. Referring specifically to the flow chart of FIG. 6, the question is there a change in steam demand is asked and a "yes" answer starts the process. If there is a steam demand decrease, the right hand portion of the flow chart is used. Where steam demand is decreased the index of performance is computed for each boiler and ΔWs steam production in pounds per hour computed for the highest index of performance. If the ΔWs of the best boiler is sufficient to meet the decrease, the target is changed and an exit is made. If not, ΔWs is computed for the next best index of performance until the demand is met.
Referring to the "yes" output of the steam demand increase block, again the ΔIP is computed for each fuel and boiler and the ΔWs is computed for the best index of performance for the purpose of increasing steam demand. Again in a manner similar to steam decrease, the same steps are made in order to meet the new demand.
As shown in FIG. 8 because of relatively long time and the regularity of the steam flows, there is ample time for bump tests and thus this allows the control algorithms to compute in advance fuel change when a new steam production level is demanded. However, as illustrated in FIG. 9, labeled method #2, if the steam flow demands change frequently the bump test cannot be made in the same manner. Thus, the flow chart of FIG. 7 is utilized when steady state operation is of extremely short duration. In essence what is done here is that changes in steam demand are actually taken with the swing fuel (such as oil) to the boilers and thereafter the base fuel flow is changed to measure its incremental efficiency while cutting back on the swing fuel in order to maintain the same steam output. Referring in detail now to FIG. 7 the efficiency computation is intiated by determining whether a change in steam production is required. If yes, swing fuel is changed to accommodate the new steam production and an incremental efficiency is computed designated η1. Thereafter, the base load fuel for the boiler is changed but since steam is maintained constant as indicated the incremental efficiency base load fuel is measured by utilizing the product of the Δ swing fuel and its efficiency as an indication of the simulated change in steam flow due to change in base fuel. The same procedure is carried out for each boiler of the system. Then fuel flows based on these efficiencies can be computed in the same manner as for method #1 in conjunction with the flow chart of FIG. 6. However, the entry, of course, will not be in response to a steam demand but made independently as indicated by the method #2 optimization entry.
Although applied to a steam distribution system the methods #1 and #2 of FIGS. 5 and 7, respectively can also be applied to a turbine generator. For such a system the efficiency is a ratio of the Δ megawatt output over the Δ throttle flow. In other words, a change in the amount of steam extracted from a single stage of a turbine will change the steam power input to the generator. Thus, changes in electrical demand or extraction flow can be met most efficiently by selecting the optimal real time cost changes in flows to the turbine. Again, an incremental real time efficiency computation is desired since unmeasurable perturbations in the turbine generator system such as heat and friction, for example, will cause hour to hour changes in efficiencies. In addition, in a turbine generator the incremental efficiency must also be calculated when the condensate is being varied.
The following example will indicate the methods of the present invention. The case is of two boilers, PB1 and PB2, where boiler 1 has oil as fuel, f1, and boiler also has oil as fuel designated f2. Also let ##EQU1## For this system the objective is to maintain steam demand at a minimum fuel cost.
A simple model for the total steam production of this system is
Ws = η1 f1 + η2 f2 (3)
Ws = Total steam produced (lb/hr)
f1 = Fuel oil to PB1 (lb/hr)
f2 = Fuel oil to PB2 (lb/hr)
The index of performance for this system must relate the cost of operation and the value of the product. Thus,
IP = Co (f1 +f2) - Cs Ws (4)
Co = Cost of fuel oil ($/lb)
Cs = Cost of purchasing steam externally ($/lb)
Ip = index of performance ($/hr)
Equation (3) describes the capability of the system to produce steam. For a change in f1 or f2 or both, the model will compute the corresponding change in steam production. As a boiler changes operating points, its efficiency parameter η will also change. Thus, η1 and η2 must be continually updated for equation (3) to be meaningful.
Equation (4) provides information concerning the cost of operation. The first term is obviously the fuel cost. The second term forces the optimization to ensure that boilers produce the amount of steam required by the process.
In order to implement this scheme, it is convenient to use the incremental form of equations (3) and (4):
MODEL Ws = η1 Δf1 + η2 Δf2 (5)
INDEX ΔIP = Co (Δf1 =Δf2) - Cs ΔWs (6)
Equation (5) computes the incremental change in steam production for an incremental change in fuel. Equation (6) computes the corresponding change in operating cost.
Given equations (5) and (6), for a required change in steam demand, to which power boiler should fuel be increased or decreased?
Assume for now that it is decided that all load swings will be taken by PB1. Then, since PB2 is being run as a base loaded unit,
Δf2 = 0
all the time. Equations (5) and (6) simplify to
ΔWs = η1 Δf1 (5a)
ΔIP1 = Co Δf1 - Cs ΔWs (6a)
Equation (5a) can be rewritten as ##EQU2## Substituting equation (5b) into equation (6a) yields ##EQU3## Equation (7a) computes the change in the cost of operation as a function of the change in steam demand if PB1 is dedicated as the swing boiler.
Now consider the same analysis with PB2 as the only boiler allowed to change operating points and PB1 operated as a base loaded unit. For this substitution equations (5) and (6) reduce to
ΔWs = η2 Δf2 (5c)
ΔIP2 = Co Δf2 - Cs ΔWs (6b)
As before, equation (5c) can be rewritten as ##EQU4## Substituting equation (5d) into equation (6b) yields Equation (7b) computes the change in the cost of operation as a function of the change in steam demand in PB2 is used as the only swing boiler.
Now consider the case where the efficiency of PB1 is greater than the efficiency of PB2, that is
η1 > η2
For this situation
ΔIP1 < ΔIP2
that is, the index of performance for PB1 is lower in value than the index of performance of PB2. More simply stated: Equation (7a) computes to a number smaller in magnitude than does equations (7b). Thus, the incremental change in operating cost for the system will be less if PB1 is the swing boiler.
Thus, for an increase in steam demand, PB1 would take the swing (Δf2 =0) since the incremental cost of increasing steam production is less costly. For a decrease in steam demand, PB2 would take the swing of (Δf1 =0) since the incremental savings of decreasing steam production is larger.
Thus the present invention provides an improved method of integrated control of the power boilers and steam distribution in the pulp and paper industry.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3908897 *||Dec 26, 1973||Sep 30, 1975||Shell Oil Co||Off-line integration of bridge and boiler controls|
|US3942327 *||Mar 25, 1974||Mar 9, 1976||Thermo Electron Corporation||Control system for external combustion engine|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4209831 *||Sep 5, 1978||Jun 24, 1980||Leeds & Northrup Company||Method and apparatus for controlling an interconnected distribution system which includes a jointly owned generating unit|
|US4271473 *||Sep 27, 1979||Jun 2, 1981||Leeds & Northrup Company||Control of parallel operated turbines in cogeneration|
|US4403293 *||Mar 6, 1981||Sep 6, 1983||Clayton Manufacturing Company||Control apparatus for use in multiple steam generator or multiple hot water generator installations|
|US4418541 *||Mar 11, 1982||Dec 6, 1983||The Babcock & Wilcox Company||Boiler loading system|
|US4421068 *||Jul 6, 1982||Dec 20, 1983||Measurex Corporation||Optimization of steam distribution|
|US4463574 *||Mar 15, 1982||Aug 7, 1984||Honeywell Inc.||Optimized selection of dissimilar chillers|
|US4486148 *||May 17, 1982||Dec 4, 1984||Michigan Consolidated Gas Company||Method of controlling a motive power and fluid driving system|
|US4559785 *||Jan 9, 1985||Dec 24, 1985||Phillips Petroleum Company||Boiler control|
|US4607325 *||Mar 30, 1984||Aug 19, 1986||Honeywell Inc.||Discontinuous optimization procedure modelling the run-idle status of plural process components|
|US4612621 *||Mar 17, 1983||Sep 16, 1986||The Babcock & Wilcox Company||Distributed system for optimizing the performance of a plurality of multi-stage steam turbines using function blocks|
|US4685072 *||Jan 2, 1985||Aug 4, 1987||The Babcock & Wilcox Company||Steam generator on-line efficiency monitor|
|US4686832 *||Apr 28, 1986||Aug 18, 1987||Miliaras Emmanuel S||Integrated fuel cleaning and power generation|
|US4745758 *||Dec 15, 1986||May 24, 1988||Westinghouse Electric Corp.||System for economic unit load distribution during process load transition|
|US4752697 *||Apr 10, 1987||Jun 21, 1988||International Cogeneration Corporation||Cogeneration system and method|
|US4864972 *||Jun 8, 1987||Sep 12, 1989||Batey John E||Boiler optimization for multiple boiler heating plants|
|US5149985 *||Oct 31, 1990||Sep 22, 1992||Nec Corporation||Two-battery supply for operating and waiting modes of a carphone|
|US5278772 *||May 6, 1992||Jan 11, 1994||Honeywell Inc.||Real-time economic load allocation|
|US5347466 *||Jul 15, 1991||Sep 13, 1994||The Board Of Trustees Of The University Of Arkansas||Method and apparatus for power plant simulation and optimization|
|US5621654 *||Apr 15, 1994||Apr 15, 1997||Long Island Lighting Company||System and method for economic dispatching of electrical power|
|US5666297 *||May 13, 1994||Sep 9, 1997||Aspen Technology, Inc.||Plant simulation and optimization software apparatus and method using dual execution models|
|US5886895 *||Jul 1, 1996||Mar 23, 1999||Kabushiki Kaisha Toshiba||Plant utility optimizing method and an optimizing system|
|US6230495 *||Nov 19, 1997||May 15, 2001||Steag Encotec And Ketek Engineering Gmbh Engergieund Umwelttechnik||Method for optimizing fossil-fueled power stations|
|US7120610 *||Feb 11, 2000||Oct 10, 2006||Ascom Hasler Mailing Systems, Inc.||Technique for effective management of resource consumption|
|US7288921 *||Jun 25, 2004||Oct 30, 2007||Emerson Process Management Power & Water Solutions, Inc.||Method and apparatus for providing economic analysis of power generation and distribution|
|US7385300||Mar 13, 2007||Jun 10, 2008||Emerson Process Management Power & Water Solutions, Inc.||Method and apparatus for determining actual reactive capability curves|
|US7474080||Sep 12, 2007||Jan 6, 2009||Emerson Process Management Power & Water Solutions, Inc.||Method and apparatus for providing economic analysis of power generation and distribution|
|US8161724||Dec 1, 2010||Apr 24, 2012||Eif Nte Hybrid Intellectual Property Holding Company, Llc||Hybrid biomass process with reheat cycle|
|US8495878||Aug 28, 2012||Jul 30, 2013||Eif Nte Hybrid Intellectual Property Holding Company, Llc||Feedwater heating hybrid power generation|
|US8596034||Mar 31, 2010||Dec 3, 2013||Eif Nte Hybrid Intellectual Property Holding Company, Llc||Hybrid power generation cycle systems and methods|
|US20040059691 *||Sep 20, 2002||Mar 25, 2004||Higgins Robert L.||Method for marketing energy-use optimization and retrofit services and devices|
|US20050285574 *||Jun 25, 2004||Dec 29, 2005||Huff Frederick C||Method and apparatus for providing economic analysis of power generation and distribution|
|US20080004721 *||Sep 12, 2007||Jan 3, 2008||Emerson Process Management Power & Water Solutions, Inc.||Method and Apparatus for Providing Economic Analysis of Power Generation and Distribution|
|US20110140453 *||Dec 1, 2010||Jun 16, 2011||Eif Nte Hybrid Intellectual Property Holding Company, Llc||Hybrid biomass process with reheat cycle|
|DE3235909A1 *||Feb 24, 1982||Feb 24, 1983||Clayton Manufacturing Co||Regelungsvorrichtung fuer dampf- bzw. heisswasser-erzeugungsanlagen mit mehreren dampf- bzw. heisswasser-erzeugern|
|DE4404272A1 *||Feb 10, 1994||Aug 11, 1994||Hitachi Ltd||Process for operating installations and system for controlling the operation of installations|
|DE4404272C2 *||Feb 10, 1994||Apr 1, 1999||Hitachi Ltd||Verfahren und Vorrichtung zum Betreiben einer Energieversogungsanlage|
|WO1982003133A1 *||Feb 24, 1982||Sep 16, 1982||Clayton Manufacturing Co||Control apparatus for use in multiple steam generator or multiple hot water generator installations|
|WO2013000733A1 *||Jun 15, 2012||Jan 3, 2013||Siemens Aktiengesellschaft||Device and method for additional oil firing for the immediate, rapid, and temporary power output increase of a coal-fired steam-electric power plant|
|WO2013163722A1 *||May 4, 2012||Nov 7, 2013||Enero Solutions Inc.||Control system for allocating steam flow through elements|
|WO2016099412A1 *||Sep 11, 2015||Jun 23, 2016||Turkiye Petrol Rafinerileri A. S. Tupras||A steam load shedding system|
|U.S. Classification||60/667, 700/38, 122/448.3, 122/448.1, 60/676, 700/36, 700/286|
|International Classification||B01J19/00, F01K3/24, F01K23/00, F01K13/02|
|Cooperative Classification||B01J19/0033, Y02E20/14, F01K3/24, F01K13/02|
|European Classification||F01K13/02, F01K3/24, B01J19/00B6|